16:01:03 #startmeeting F26-blocker-review 16:01:03 Meeting started Mon Jun 5 16:01:03 2017 UTC. The chair is roshi. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:01:03 Useful Commands: #action #agreed #halp #info #idea #link #topic. 16:01:03 The meeting name has been set to 'f26-blocker-review' 16:01:03 #meetingname F26-blocker-review 16:01:03 The meeting name has been set to 'f26-blocker-review' 16:01:03 #topic Roll Call 16:01:12 who's around for some blocker review!? 16:01:14 .hello jkurik 16:01:15 jkurik: jkurik 'Jan Kurik' 16:01:18 ahoyhoy 16:01:20 .hello adamwill 16:01:22 adamw: adamwill 'Adam Williamson' 16:01:26 hello 16:01:59 hi again nathan 16:02:06 * roshi has coffee and something to eat - which is win :D 16:02:24 #chair jkurik adamw tflink kparal sgallagh 16:02:24 Current chairs: adamw jkurik kparal roshi sgallagh tflink 16:02:39 * jkurik has no coffee, no food :( 16:02:48 * adamw has coffee and shreddies 16:02:52 breakfast of champions 16:03:08 * roshi has 3 beer brats, and some sriracha mayo 16:03:15 * pschindl_wfh is here 16:03:22 #chair pschindl_wfh 16:03:22 Current chairs: adamw jkurik kparal pschindl_wfh roshi sgallagh tflink 16:03:31 #topic Introduction 16:03:31 Why are we here? 16:03:31 #info Our purpose in this meeting is to review proposed blocker and nice-to-have bugs and decide whether to accept them, and to monitor the progress of fixing existing accepted blocker and nice-to-have bugs. 16:03:35 #info We'll be following the process outlined at: 16:03:37 #link https://fedoraproject.org/wiki/QA:SOP_Blocker_Bug_Meeting 16:03:40 #info The bugs up for review today are available at: 16:03:42 #link http://qa.fedoraproject.org/blockerbugs/current 16:03:45 #info The criteria for release blocking bugs can be found at: 16:03:47 #link https://fedoraproject.org/wiki/Fedora_26_Alpha_Release_Criteria 16:03:50 #link https://fedoraproject.org/wiki/Fedora_26_Beta_Release_Criteria 16:03:53 #link https://fedoraproject.org/wiki/Fedora_26_Final_Release_Criteria 16:04:02 * kparal is here 16:04:16 #info there's no proposals for Beta 16:04:25 and we kinda already know the status of the libdb thing 16:04:27 \o/ 16:04:31 so moving directly into Final proposals' 16:04:40 someone check the duct tape on kparal 16:04:44 7 proposed blockers 16:05:01 he's trying to tap out 'BetaBlocker' with his eyebrows, i can see it 16:05:01 first up 16:05:02 #topic (1454854) DNF progress.start() broken with 2.5.0, breaking dnfdaemon (and dnfdragora) 16:05:05 #link https://bugzilla.redhat.com/show_bug.cgi?id=1454854 16:05:08 lol 16:05:10 #info Proposed Blocker, dnfdaemon, ON_QA 16:06:01 the question of what's the 'default' package manager on KDE seems a bit complex ATM 16:06:06 but dnfdragora at least seems to be in the running 16:06:17 * roshi for sure doesn't know wha tthe default is on KDE 16:07:22 I did ask "Beyond this: what's the actual intended shape of package management and 16:07:22 updates on KDE now? Is dnfdragora supposed to be responsible for all of 16:07:22 it, or is...some other thing...supposed to handle updating, and 16:07:22 dnfdragora is only supposed to be there for installing / removing?" on the KDE list a while back 16:07:25 aha, just the person 16:07:40 kevin, can you clarify what exactly is the 'default package manager' for KDE right now? 16:07:47 welcome Kevin_Kofler :) 16:09:47 when I run "Software center" from KDE menu, it runs /usr/bin/plasma-discover 16:10:05 +1 assuming dnfdragora is the default manager 16:10:10 if not, we can retract 16:10:11 which seems to have the ability to update sw and install/remove packages 16:10:21 it will be solved in a week anyway, hopefully 16:10:28 yeah 16:10:35 looks like there's a fix already 16:10:55 since we don't know the default, what say we punt and see if it resolves before we meet again? 16:11:05 i'd say ask for clarification in the bug, and punt 16:11:19 adamw: +1 16:12:13 As I understand it, both dnfdragora and Plasma Discover should be shipped by default, Apper should not anymore. 16:12:32 proposed #agreed - RHBZ#1454854 - Punt - It's not 100% clear what is the default package manager for KDE, so we can't determine blocker status for this issue. 16:12:38 patch 16:12:39 dnfdragora manages packages at package level, Plasma Discover is like GNOME Software (AppStream-based). 16:12:51 I can ask on #fedora-kde. 16:12:52 proposed #agreed - RHBZ#1454854 - Punt - It's not 100% clear what is the default package manager for KDE, so we can't determine blocker status for this issue at this time. 16:13:07 that'd be great Kevin_Kofler 16:13:20 ack 16:13:23 circle back on this or wait? 16:13:26 * roshi is fine with either 16:13:31 ack 16:13:47 ack to the proposed, or to waiting? 16:13:48 I asked there, let's see if anybody knows more than me. 16:14:15 okay. we may even have to amend the criteria to say default package manager(s) or something, then. 16:14:20 rdieter probably knows for sure, but he's not online now. 16:14:24 So maybe punt for now. 16:14:27 and consider whether it's okay for just *one* to work, or if they all have to work... 16:15:08 I'd say all. same as all default apps 16:15:14 yeah 16:15:40 but that would raise the question why kde needs 2 things (if I understand correctly) 16:16:08 Because Discover manages only "applications", not packages. 16:16:16 You see only what is in AppStream. 16:16:21 yeah, same reason some people install yumex or whatever on Workstation. 16:16:25 It also does not show the dependencies etc. 16:16:45 and dnfdragora does which bits that dnf itself doesn't do? 16:16:50 By the way, dnfdragora should work great on Workstation too, there is even a GTK+ backend. (It uses the libyui abstraction.) 16:17:00 dnfdragora is a GUI for DNF. :-) 16:17:06 so, i'm still fine with a punt 16:17:08 ah 16:17:12 (and an ncurses TUI too, libyui gives you that for free) 16:17:18 so, acks to the proposed? 16:17:44 It's a frontend for dnfdaemon which uses the DNF Python API. 16:17:48 .fas linuxmodder 16:17:48 linuxmodder: linuxmodder 'Corey W Sheldon' 16:17:58 proposed #agreed - RHBZ#1454854 - Punt - It's not 100% clear what is the default package manager for KDE, so we can't determine blocker status for this issue at this time. 16:18:45 ack 16:18:46 ack again 16:18:49 ack 16:18:53 #agreed - RHBZ#1454854 - Punt - It's not 100% clear what is the default package manager for KDE, so we can't determine blocker status for this issue at this time. 16:19:02 ah 16:19:10 can someone secretarialize? 16:19:18 * roshi forgot to ask for that earlier 16:19:41 #topic (1452866) FreeIPA upgrade script requires network to be up, but network is not up during upgrade when using dnf system-upgrade 16:19:44 #link https://bugzilla.redhat.com/show_bug.cgi?id=1452866 16:19:45 i can 16:19:46 #info Proposed Blocker, freeipa, NEW 16:19:50 thanks adamw 16:20:12 so i've been looking into this one some more, but it's still not yet clear precisely what the impact is 16:20:15 i need to run some more tests 16:20:35 right now it's clear the upgrade process doesn't really work properly at all, but the server does still work to *some* extent despite the failed upgrade script 16:20:48 we need to nail down some details on what the upgrade script failing actually breaks 16:22:00 so punt? 16:22:16 yeah, for no 16:22:17 for now 16:22:23 sounds like a valid blocker to me but like adamw I'd have to test some more 16:22:30 so I'm +1 punt 16:23:31 works for me 16:23:36 * roshi was reading 16:24:20 proposed #agreed - RHBZ#1452866 - Punt - It's still not 100% clear what the impact of this issue is, and we'll wait to determine blocker status until we have more information. 16:24:29 ack 16:24:29 ack 16:24:39 ack 16:24:40 #agreed - RHBZ#1452866 - Punt - It's still not 100% clear what the impact of this issue is, and we'll wait to determine blocker status until we have more information. 16:24:48 #topic (1450638) [abrt] gnome-font-viewer: gtk_header_bar_pack(): gnome-font-viewer killed by signal 11 16:24:51 #link https://bugzilla.redhat.com/show_bug.cgi?id=1450638 16:24:53 #info Proposed Blocker, gnome-font-viewer, ON_QA 16:25:52 I think I reproduced this one as well, very simple 16:25:54 +1 16:26:12 not something I use but can see it being an issue 16:26:31 sure, seems pretty 'basic functionality' to me 16:26:32 +1 16:26:37 +1 16:26:43 +1 16:26:46 it seems to be already fixed anyway 16:27:21 in what package ? jkurik 16:28:01 proposed #agreed - RHBZ#1450638 - AcceptedBlocker - This bug is a violation of the following Final criterion: "All applications that can be launched using the standard graphical mechanism of a release-blocking desktop after a default installation of that desktop must start successfully and withstand a basic functionality test." 16:28:23 linuxmodder: I mean it is waiting for karma https://bodhi.fedoraproject.org/updates/FEDORA-2017-1d5ee3c3a7 16:28:35 jkurik, will check hat out 16:28:51 roshi: ack 16:29:09 ack 16:29:09 took 2 weeks to get a +3 yikes 16:29:26 ack 16:29:26 Hmmm, now that I think of it, isn't the dnfdragora bug a blocker under this criterion even if we decide that Discover is the "default package manager" on KDE Plasma? 16:29:46 #agreed - RHBZ#1450638 - AcceptedBlocker - This bug is a violation of the following Final criterion: "All applications that can be launched using the standard graphical mechanism of a release-blocking desktop after a default installation of that desktop must start successfully and withstand a basic functionality test." 16:29:53 he's correct 16:29:59 Kevin_Kofler: seems like 16:30:12 we should know our criteria better :) 16:30:14 Kevin_Kofler: fair point 16:30:15 you can launch kde form sddm too so yeah makes sense 16:30:18 * roshi still wonders why a GUI for dnf is shipped, but hey, it's not my spin :) 16:30:23 where's that dog-in-a-tie meme 16:30:23 so we can go back to it and ack it 16:30:28 it should technically be fixed 16:30:47 the dnf guys fixed dnfdaemon, which I merged and released 16:31:06 #topic (1454854) DNF progress.start() broken with 2.5.0, breaking dnfdaemon (and dnfdragora) 16:31:06 then they reverted their change, which besser82 applied downstream to dnfdaemon 16:31:06 #link https://bugzilla.redhat.com/show_bug.cgi?id=1454854 16:31:06 fixing it again 16:31:06 #info Proposed Blocker, dnfdaemon, ON_QA 16:31:07 ok, back to the bug in question 16:31:17 under that criterion, I can be +1 16:31:21 if it ships by default 16:31:37 +1 16:32:01 yep, +1 too 16:32:11 +1 makes sense 16:32:39 +1 16:32:50 proposed #agreed - RHBZ#1454854 - AcceptedBlocker - Because this ships by default on KDE, it's a violation of the following criterion: "All applications that can be launched using the standard graphical mechanism of a release-blocking desktop after a default installation of that desktop must start successfully and withstand a basic functionality test." 16:32:55 ack 16:32:57 +1 FE 16:33:01 ack 16:33:03 ack 16:33:12 afaik it does not ship on any release blocking media 16:33:18 KDE 16:33:27 it ships in KDE, LxQt, and Cinnamon 16:33:28 dgilmore: it's on KDE. 16:33:31 at least KDE is blocking 16:33:36 okay 16:33:46 #agreed - RHBZ#1454854 - AcceptedBlocker - Because this ships by default on KDE, it's a violation of the following criterion: "All applications that can be launched using the standard graphical mechanism of a release-blocking desktop after a default installation of that desktop must start successfully and withstand a basic functionality test." 16:34:01 #topic (1451914) [abrt] gnome-shell: GjsMaybeOwned::trace(): gnome-shell killed by signal 6 16:34:04 #link https://bugzilla.redhat.com/show_bug.cgi?id=1451914 16:34:07 #info Proposed Blocker, gnome-shell, NEW 16:35:01 +1 under the same criterion 16:35:26 I think we can fit it to the default panel functionality 16:35:28 hum 16:35:32 don't we have another report of this? 16:35:38 i'd say the 'data loss' criterion 16:36:02 see comment 18 16:36:15 data loss is even better 16:36:26 i was thinking of https://bugzilla.redhat.com/show_bug.cgi?id=1450639 16:36:42 +1, watching a movie is an important activity which should not be interrupted by gnome-shell crash 16:36:58 heh 16:37:17 the first one has more crashes in FAF 16:37:56 so, thinking about it, i can be +1 to any commonly-encountered gnome-shell crash so long as gnome-shell crashing takes down the desktop under Wayland 16:38:01 they keep saying they're gonna fix that but they haven't yet 16:38:07 so, this seems to have lots of reports, so +1 16:38:12 +1 16:38:34 under the default functionality or data-loss criterion? 16:38:44 I'm unsure of what "lost all data" means 16:39:47 * roshi assumes session data 16:39:53 open docs and whatnot 16:40:09 yea 16:40:11 i think so 16:40:31 i mean, we could even call it 'basic functionality' for gnome-shell, really 16:40:31 I would prefer the "default functionality" one, just because it fits better according to my feeling 16:40:32 which, to me, is *technically* data loss, but not in the "oh noes!" tone it tends to be seen as 16:40:40 it shouldn't be possible to crash the damn shell by playing a video 16:40:48 yeah, true 16:40:53 jkurik: which one do you mean specifically? 16:41:01 no video has killed my i3wm setup! 16:41:21 All applications that can be launched using the standard graphical mechanism of a release-blocking desktop after a default installation of that desktop must start successfully and withstand a basic functionality test. 16:41:25 is my vote 16:41:47 adamw: yeah, this one ^^^ 16:41:53 sure 16:41:54 roshi: thanks, you were faster 16:41:58 np 16:42:00 we can even mention multiple criteria 16:42:02 no law against it 16:42:03 :P 16:42:05 already had it open 16:42:47 proposed #agreed - RHBZ#1451914 - AcceptedBlocker - This bug is a violation of a few criteria, but notably "All applications that can be launched using the standard graphical mechanism of a release-blocking desktop after a default installation of that desktop must start successfully and withstand a basic functionality test." 16:43:02 ack 16:43:05 ack 16:43:28 ack 16:43:29 #agreed - RHBZ#1451914 - AcceptedBlocker - This bug is a violation of a few criteria, but notably "All applications that can be launched using the standard graphical mechanism of a release-blocking desktop after a default installation of that desktop must start successfully and withstand a basic functionality test." 16:43:40 #topic (1455673) GNOME Software should make sure system is current on updates before upgrading to new release 16:43:43 with roshi on this 16:43:43 #link https://bugzilla.redhat.com/show_bug.cgi?id=1455673 16:43:46 #info Proposed Blocker, gnome-software, NEW 16:44:00 oh look 16:44:13 we just talked about this 16:44:32 lol 16:44:55 is mattdm travelling through time? 16:45:02 probably 16:45:07 * roshi suspects he's a bot as well 16:45:24 so many people working in fedoraland don't seem to be human in their efforts 16:45:28 should not this be a Feature request, instead of a blocker bug ? 16:45:32 the bots wars were already won by adamw 16:45:49 nah, the puiterwijk cloud has me beat 16:45:53 seems legit 16:46:01 in general this feels a bit feature request-y, yeah 16:46:03 this doesn't really violate any criteria 16:46:08 but there is the context of the libdb issue 16:46:27 sticking strictly to the intent of this meeting, I have to say -1 16:46:37 yes, I see the context as well... but ... 16:46:56 since that's not really the purpose of the meeting, and I don't think we should use this meeting as the means to fix this particular issue 16:47:00 we can't really mandate this, I think. we can block upgrade issues, or we can demand some solution. be it written instructions, ehm ehm... 16:47:31 -1 (sorry mattdm) 16:47:36 FESCO could 16:48:22 yeah 16:48:39 right, i think the appropriate thing for this would be for FESCo to use its blocker wand or make it a prioritized bug 16:48:42 I agree with the bug (I honestly thought that would be the default) 16:48:48 * kparal also notes https://bugzilla.redhat.com/show_bug.cgi?id=1336435 16:49:10 let's propose it for a prioritized bug 16:49:14 anyone remembers how? 16:49:53 I SUMMON THEE FESCO! 16:49:55 * roshi waits 16:50:01 er 16:50:04 * sgallagh pops in 16:50:05 I SUMMON THEE, FESCO! 16:50:13 just needed the comma 16:50:18 did you sacrifice the rubber chicken? 16:50:18 should I dupe it to 1336435 right away? 16:50:30 it's a dupe, so I'd say yeah 16:50:32 I did 16:50:40 +1 to dup 16:50:42 https://fedoraproject.org/wiki/Fedora_Program_Management/Prioritized_bugs_and_issues_-_the_process 16:50:44 complete with vegan vegetable blood 16:52:40 I am +1 to mark both bugs as duplicates and then propose it as Prioritized (I can do the proposal then) 16:52:45 ok, I just duped it and proposed it as prioritized 16:52:52 proposed #agreed - RHBZ#1455673 - RejectedBlocker - This bug doesn't violate any criteria, and is therefore non-blocking. While we agree with the intent, we'll propose this as a PrioritizedBug for review (as that seems the more appropriate venue). 16:53:02 thanks kparal 16:53:05 ack 16:53:21 ack 16:53:48 ack 16:53:54 #agreed - RHBZ#1455673 - RejectedBlocker - This bug doesn't violate any criteria, and is therefore non-blocking. While we agree with the intent, we'll propose this as a PrioritizedBug for review (as that seems the more appropriate venue). 16:54:02 #topic (1433884) SELinux is preventing (fwupd) from 'mounton' accesses on the directory /var/lib/fwupd. 16:54:05 #link https://bugzilla.redhat.com/show_bug.cgi?id=1433884 16:54:07 #info Proposed Blocker, selinux-policy, NEW 16:54:31 btw: it gets immediatelly rejected as Prioritized bug by robatino 16:54:46 i don't think he can do that... 16:54:57 I mean the #1336435 - why ? 16:55:08 "orry about that. I just wanted to subscribe and didn't get the usual "mid-air 16:55:08 collision" prompt." 16:55:31 ah, ok... 16:55:40 I figured 16:55:48 re-added it 16:55:55 what're we talking about now? 16:55:57 last bug? 16:56:16 yeah. let's go to the current one 16:56:29 so, no proposal, no information when this happens 16:57:10 well, we had 16:57:11 https://bugzilla.redhat.com/show_bug.cgi?id=1429341https://bugzilla.redhat.com/show_bug.cgi?id=1429341 16:57:19 which this is listed as a 'possible dupe' of, at two removes 16:57:39 and indeed this was filed with 3.13.1-246 (while the fix was in 3.13.1-247)... 16:58:33 right, looks like duplicates and already fixed 16:58:49 so, punt and ask to re-test 16:58:56 sure 16:59:28 for the current selinux bug? 16:59:42 * roshi lost track of who was talking about what there 16:59:47 yes, for the 1433884 17:00:41 proposed #agreed - RHBZ#1433884 - Punt - It looks like this is already fixed. Please retest and update the bug. 17:01:06 ack 17:01:06 patch 17:01:11 ack 17:01:20 go for it jkurik you have chair 17:01:39 proposed #agreed - RHBZ#1433884 - Punt - It looks like this is already fixed in build 3.13.1-247. Please retest and update the bug. 17:01:46 ack 17:01:51 just to give a bit more info where to look for the fix 17:02:14 you can go ahead and to the #agreed after the acks 17:02:22 #agreed - RHBZ#1433884 - Punt - It looks like this is already fixed in build 3.13.1-247. Please retest and update the bug. 17:02:35 #topic (1449643) Running groupadd produces AVC denials about net_admin and write to system_dbusd_var_run_t 17:02:38 #link https://bugzilla.redhat.com/show_bug.cgi?id=1449643 17:02:40 #info Proposed Blocker, selinux-policy, POST 17:04:08 if it happens during an install of workstation, +1 17:04:35 yeah 17:04:43 I saw it several times during dnf install. but one of my comments says it happened during installation, so I guess I have to trust myself 17:05:01 dangerous game, that 17:05:05 trusting yourself 17:05:31 proposed #agreed - RHBZ#1449643 - AcceptedBlocker - This bug violates the following criterion: "There must be no SELinux denial notifications or crash notifications on boot of or during installation from a release-blocking live image, or at first login after a default install of a release-blocking desktop." 17:05:38 ack 17:06:30 I wonder if I might have installed something prior to running installation 17:07:00 * kparal shrugs 17:07:08 eh, we can downgrade it later 17:07:09 ack 17:07:19 #agreed - RHBZ#1449643 - AcceptedBlocker - This bug violates the following criterion: "There must be no SELinux denial notifications or crash notifications on boot of or during installation from a release-blocking live image, or at first login after a default install of a release-blocking desktop." 17:07:25 #topic Open Floor 17:07:28 that's it for blockers 17:07:44 I don't see a reason to go through proposed FEs since we're not close to Final Freeze 17:07:51 * kparal nods 17:08:03 * jkurik agrees 17:09:01 agreed 17:09:06 anything else? 17:09:13 not from me 17:09:19 * roshi sets the fuse... 17:09:21 3... 17:09:25 more coffee for roshi 17:09:30 yup yup 17:09:42 couple tablespoons of butter, some cinnamon 17:11:09 2... 17:11:11 1... 17:11:15 thanks for coming folks! 17:11:18 #endmeeting